Comparison of Pulmonary Artery, Rectal, and Tympanic Membrane Temperatures in Adult Intensive Care Unit Patients Anne Milewski, BS, RN*, Kevin L.
Ferguson, MD**,
Thomas E.
Terndrup, MD, FACEP†
Tympanic thermometry using infrared thermography technology offers a noninvasive, rapid temperature measurement tool which may be useful for selected intensive care unit (ICU) patients. Multiple comparisons of pulmonary artery catheter (PAC), rectal (R), and tympanic membrane (TM) temperatures were performed in nine adult ICU patients using PAC temperature as the gold standard. The correlation between R (r=0.93) and PAC was significantly better than TM (r=0.74) temperatures. However, PAC (37.2 ± 0.06°C ; mean ± SEM) and TM (37.1 ± 0.08°C) temperatures were not significantly different, whereas R (37.6 ± 0.07°C) was significantly warmer than both (P PAC > TM temperatures during most time intervals.
gaussian
FIG. 2. Histogram of pulmonary artery minus tympanic membrane (bottom) temperatures.
rectal
(top)
or
Discussion 23°C during the study period. Indications for insertion of the PAC were perioperative monitoring, septic shock, and intravascular volume status determination. Pulmonary artery catheter (37.2’ ± 0.06°C; mean + SEM) and TM (37.1 ° ± 0.08°C) temperatures were not significantly different. However, R (37.6° ± 0.07°C) was significantly warmer than both PAC and TM temperatures (P < 0.05). Generous ranges were observed for PAC (34.5 40°C), R (35.4° - 40.4°C), and TM (34.1 ° - 40°C) temperatures. Differences between PAC and either R or TM temperatures over selected temperature ranges varied somewhat (Table). The range of differences was smaller for R (-1.10 to -0.46) than for TM (-.35 to 0.22) temperatures. Correlation between PAC and R (r=0.93) temperature was significantly better than to TM (r=0.74; Figure 1). Using linear regression, the relationship between PAC and R temperature is PAC 0.83R + 6.79 and for TM temperatures, PAC = 0.88TM + 4.49. The 95% confidence intervals for the slopes of the respective linear regression lines are for R (0.78 to 0.89) and for TM (0.76 to 0.99). The absolute difference in the recorded temperatures tended to be smaller and more often negative for PAC-RT (Figure 2, top), while PAC-TMT showed a wider and more
In adult ICU patients, both R and TM temperatures show linear relationship with the PAC temperatures using regression analysis. While TM temperatures demonstrate a wider variation than the R temperatures and correlated less well with PAC temperatures, there was no significant difference between mean PAC and TM temperatures. a
°
-
=
FIG. 4. Individual time versus pulmonary and tympanic membrane temperatures.
artery catheter, rectal, 15
Differences between PAC and R or TM temperatures were smaller for PAC-TM than for PAC-R. This relationship was present throughout the recorded temperature range. When individual patient’s fever trends were plotted against time, the correlation between the three devices continued, with the temperature trend lines roughly parallel. None of the devices failed to detect a fever in any patient, though at any given point there may have been a discrepancy in recorded temperatures. Similarly, none of the devices reported a fever not confirmed by other devices using 38.0°C as the fever cutoff for each device. This study supports several previous findings of the relationship between the different methods of temperature measurement. The tendency for R to be higher than the P AC,7,8 and the tendency for the TMT to be cooler than the PAC and R temperature has been reported.9-’‘ Two explanations for site and thermometer dependency seem likely. First, the devices have varying sensitivity and precision in temperature measurement. Second, regional body temperatures vary significantly. 11,11,14 The TM temperature may more closely reflect actual brain temperature, since the blood supply to the TM comes from the hypothalamus. If the TM thermometer is accurate, it may provide better information on the temperature perceived by the brain’s &dquo;thermostat&dquo; than temperature at other body sites. While this study involved only a small number of patients, multiple comparisons of PAC, R, and TM temperatures were obtained over a broad temperature range. Also, we limited data from individual study patients to a maximum of 48 hours and examined individual temperature trends to see if one patient’s data would greatly influence the group data. The ICU nursing staff was not blinded to temperature measurement and the greater scatter observed for TM temperatures may have been due to operator or device performance. The data obtained was limited to adult patients and may not apply to pediatric ICU patients. Regional body temperature differences may be smaller in
pediatric patients.’S Conclusions The
performance
of rectal and
tympanic
membrane
temperatures was similar in adult patients, compared to pulmonary artery catheter thermistor readings. Although both rectal and tympanic membrane temperatures correlated linearly with pulmonary artery catheter temperatures, the latter was associated with
a
wider data variance.
Acknowledgements The authors wish 16
to
express
our
thanks to the 8C ICU
nursing staff at SUNY Health Science Center @ Syracuse for their assistance in conducting this study. References 1. Cranston WJ. Temperature regulation. Brit MedJ 1966; 2:69-75. 2. Bates DW, Cook EF, Goldman L, et al. Predicting bacteremia in hospitalized patients: a prospectively validated model. Ann Int Med 1990; 113:495-500. 3. Temdrup TE, Allegra JR, Kealy JA. A comparison of oral, rectal, and tympanic membrane-derived temperature changes after ingestion of liquids and smoking. Am J Emerg Med 1989; 7:150-4. 4. Gruber PA. Changes in cardiac rate associated with the use of the rectal thermometer in the patient with acute myocardial infarction. Heart and Lung 1974; 3:288-92. 5. Earnest DL, Fletcher GF. Danger of rectal examination of patients with acute myocardial infarction—fact or fiction? N Engl J Med 1969; 281:238-41. 6. Shinozaki T, Deane R, Perkins FM. Infrared tympanic thermometer: evaluation of a new clinical thermometer. Crit Care Med 1988; 16:148-50. 7. Ilsley AH, Rutten AJ, Runciman WB. An evaluation of body temperature measurement. Anesth Intens Care 1983; 11:31-9. 8. Shiraki K, Konda N, Sagawa S. Esophageal and tympanic temperature responses to core blood temperature changes during hyperthermia. J Appl Physiol 1986; 61:98-102. 9. Muma BK, Treolar DJ, Wurmlinger K, et al. Comparison of rectal, axillary, and tympanic membrane temperatures in infants and young children. Ann Emerg Med 1991; 20:41-4. 10. Rhoads FA, Grandner J. Assessment of an aural infrared sensor for body temperature measurement in children. Clin Pediatr 1990; 29:112-5. 11. Ros SP. Evaluation of a tympanic membrane thermometer in an outpatient clinical setting. Ann Emerg Med 1989; 18:1004-6. 12. Cranston WI, Gerbrandy J, Snell ES. Oral, rectal and eosophageal temperatures and some factors affecting them in man. J Physiol 1954; 126:347-58. 13. DuBois EF. The many different temperatures of the human body and its parts. Western J Surg 1951; 59:476-90. 14. Molnar GW, Read RC. Studies during open-heart surgery on the special characteristics of rectal temperature. J Appl Physiol 1974; 36:333. 15. Johnson, KJ, Bhatia P, Bell EF: Infrared thermometry of newborn infants. Pediatrics 1991; 87:34-38.
Ferguson, MD**,
Thomas E.
Terndrup, MD, FACEP†
Tympanic thermometry using infrared thermography technology offers a noninvasive, rapid temperature measurement tool which may be useful for selected intensive care unit (ICU) patients. Multiple comparisons of pulmonary artery catheter (PAC), rectal (R), and tympanic membrane (TM) temperatures were performed in nine adult ICU patients using PAC temperature as the gold standard. The correlation between R (r=0.93) and PAC was significantly better than TM (r=0.74) temperatures. However, PAC (37.2 ± 0.06°C ; mean ± SEM) and TM (37.1 ± 0.08°C) temperatures were not significantly different, whereas R (37.6 ± 0.07°C) was significantly warmer than both (P PAC > TM temperatures during most time intervals.
gaussian
FIG. 2. Histogram of pulmonary artery minus tympanic membrane (bottom) temperatures.
rectal
(top)
or
Discussion 23°C during the study period. Indications for insertion of the PAC were perioperative monitoring, septic shock, and intravascular volume status determination. Pulmonary artery catheter (37.2’ ± 0.06°C; mean + SEM) and TM (37.1 ° ± 0.08°C) temperatures were not significantly different. However, R (37.6° ± 0.07°C) was significantly warmer than both PAC and TM temperatures (P < 0.05). Generous ranges were observed for PAC (34.5 40°C), R (35.4° - 40.4°C), and TM (34.1 ° - 40°C) temperatures. Differences between PAC and either R or TM temperatures over selected temperature ranges varied somewhat (Table). The range of differences was smaller for R (-1.10 to -0.46) than for TM (-.35 to 0.22) temperatures. Correlation between PAC and R (r=0.93) temperature was significantly better than to TM (r=0.74; Figure 1). Using linear regression, the relationship between PAC and R temperature is PAC 0.83R + 6.79 and for TM temperatures, PAC = 0.88TM + 4.49. The 95% confidence intervals for the slopes of the respective linear regression lines are for R (0.78 to 0.89) and for TM (0.76 to 0.99). The absolute difference in the recorded temperatures tended to be smaller and more often negative for PAC-RT (Figure 2, top), while PAC-TMT showed a wider and more
In adult ICU patients, both R and TM temperatures show linear relationship with the PAC temperatures using regression analysis. While TM temperatures demonstrate a wider variation than the R temperatures and correlated less well with PAC temperatures, there was no significant difference between mean PAC and TM temperatures. a
°
-
=
FIG. 4. Individual time versus pulmonary and tympanic membrane temperatures.
artery catheter, rectal, 15
Differences between PAC and R or TM temperatures were smaller for PAC-TM than for PAC-R. This relationship was present throughout the recorded temperature range. When individual patient’s fever trends were plotted against time, the correlation between the three devices continued, with the temperature trend lines roughly parallel. None of the devices failed to detect a fever in any patient, though at any given point there may have been a discrepancy in recorded temperatures. Similarly, none of the devices reported a fever not confirmed by other devices using 38.0°C as the fever cutoff for each device. This study supports several previous findings of the relationship between the different methods of temperature measurement. The tendency for R to be higher than the P AC,7,8 and the tendency for the TMT to be cooler than the PAC and R temperature has been reported.9-’‘ Two explanations for site and thermometer dependency seem likely. First, the devices have varying sensitivity and precision in temperature measurement. Second, regional body temperatures vary significantly. 11,11,14 The TM temperature may more closely reflect actual brain temperature, since the blood supply to the TM comes from the hypothalamus. If the TM thermometer is accurate, it may provide better information on the temperature perceived by the brain’s &dquo;thermostat&dquo; than temperature at other body sites. While this study involved only a small number of patients, multiple comparisons of PAC, R, and TM temperatures were obtained over a broad temperature range. Also, we limited data from individual study patients to a maximum of 48 hours and examined individual temperature trends to see if one patient’s data would greatly influence the group data. The ICU nursing staff was not blinded to temperature measurement and the greater scatter observed for TM temperatures may have been due to operator or device performance. The data obtained was limited to adult patients and may not apply to pediatric ICU patients. Regional body temperature differences may be smaller in
pediatric patients.’S Conclusions The
performance
of rectal and
tympanic
membrane
temperatures was similar in adult patients, compared to pulmonary artery catheter thermistor readings. Although both rectal and tympanic membrane temperatures correlated linearly with pulmonary artery catheter temperatures, the latter was associated with
a
wider data variance.
Acknowledgements The authors wish 16
to
express
our
thanks to the 8C ICU
nursing staff at SUNY Health Science Center @ Syracuse for their assistance in conducting this study. References 1. Cranston WJ. Temperature regulation. Brit MedJ 1966; 2:69-75. 2. Bates DW, Cook EF, Goldman L, et al. Predicting bacteremia in hospitalized patients: a prospectively validated model. Ann Int Med 1990; 113:495-500. 3. Temdrup TE, Allegra JR, Kealy JA. A comparison of oral, rectal, and tympanic membrane-derived temperature changes after ingestion of liquids and smoking. Am J Emerg Med 1989; 7:150-4. 4. Gruber PA. Changes in cardiac rate associated with the use of the rectal thermometer in the patient with acute myocardial infarction. Heart and Lung 1974; 3:288-92. 5. Earnest DL, Fletcher GF. Danger of rectal examination of patients with acute myocardial infarction—fact or fiction? N Engl J Med 1969; 281:238-41. 6. Shinozaki T, Deane R, Perkins FM. Infrared tympanic thermometer: evaluation of a new clinical thermometer. Crit Care Med 1988; 16:148-50. 7. Ilsley AH, Rutten AJ, Runciman WB. An evaluation of body temperature measurement. Anesth Intens Care 1983; 11:31-9. 8. Shiraki K, Konda N, Sagawa S. Esophageal and tympanic temperature responses to core blood temperature changes during hyperthermia. J Appl Physiol 1986; 61:98-102. 9. Muma BK, Treolar DJ, Wurmlinger K, et al. Comparison of rectal, axillary, and tympanic membrane temperatures in infants and young children. Ann Emerg Med 1991; 20:41-4. 10. Rhoads FA, Grandner J. Assessment of an aural infrared sensor for body temperature measurement in children. Clin Pediatr 1990; 29:112-5. 11. Ros SP. Evaluation of a tympanic membrane thermometer in an outpatient clinical setting. Ann Emerg Med 1989; 18:1004-6. 12. Cranston WI, Gerbrandy J, Snell ES. Oral, rectal and eosophageal temperatures and some factors affecting them in man. J Physiol 1954; 126:347-58. 13. DuBois EF. The many different temperatures of the human body and its parts. Western J Surg 1951; 59:476-90. 14. Molnar GW, Read RC. Studies during open-heart surgery on the special characteristics of rectal temperature. J Appl Physiol 1974; 36:333. 15. Johnson, KJ, Bhatia P, Bell EF: Infrared thermometry of newborn infants. Pediatrics 1991; 87:34-38.
